FTIR-BASED QUANTIFICATION OF THERMALLY INDUCED TRANS FATTY ACIDS IN COMMERCIAL COTTONSEED OILS: A SPECTROSCOPIC AND CHEMOMETRIC APPROACH

ABSTRACT

Trans fatty acids (TFAs) are widely recognized as dietary lipids with detrimental effects on human health. This study presents a rapid, non-destructive approach for quantifying TFAs in commercial cottonseed oil samples using Fourier-transform infrared spectroscopy (FTIR) combined with chemometric modeling. Three representative oils from Uzbekistan were subjected to thermal treatment at 100 °C, 180 °C, and 250 °C for durations of up to 180 minutes to simulate domestic and industrial cooking practices. FTIR spectra were analyzed in the 900–990 cm⁻¹ region, with the absorption band at 966 cm⁻¹ used as a diagnostic marker for trans double bonds. A robust calibration model (R² = 0.99, RMSE = 0.9%) was developed using derivative spectral preprocessing. The results revealed a significant temperature- and time-dependent increase in TFA content, particularly at 250 °C, where concentrations exceeded 3% after 180 minutes. This research confirms that excessive heating dramatically elevates TFA levels, underscoring the health risks associated with prolonged high-temperature cooking. The method proposed here offers a practical tool for monitoring TFA formation in oils without requiring complex sample preparation or chromatographic instrumentation.

АННОТАЦИЯ

Трансжирные кислоты (ТЖК) широко известны как пищевые липиды, оказывающие пагубное воздействие на здоровье человека. В этом исследовании представлен быстрый, неразрушающий подход к количественному определению ТЖК в коммерческих образцах хлопкового масла с использованием инфракрасной спектроскопии с преобразованием Фурье (ИКФУ) в сочетании с хемометрическим моделированием. Три репрезентативных масла из Узбекистана были подвергнуты термической обработке при 100 °C, 180 °C и 250 °C в течение продолжительности до 180 минут для имитации бытовых и промышленных методов приготовления пищи. Спектры ИКФУ анализировались в области 900–990 см⁻¹, при этом полоса поглощения при 966 см⁻¹ использовалась в качестве диагностического маркера для транс двойных связей. Надежная калибровочная модель (R² = 0,99, RMSE = 0,9%) была разработана с использованием производной спектральной предварительной обработки. Результаты показали значительное увеличение содержания TFA в зависимости от температуры и времени, особенно при 250°C, где концентрации превысили 3% через 180 минут. Это исследование подтверждает, что чрезмерное нагревание резко повышает уровень TFA, подчеркивая риски для здоровья, связанные с длительной высокотемпературной готовкой. Предложенный здесь метод предлагает практический инструмент для мониторинга образования TFA в маслах без необходимости сложной подготовки образцов или хроматографического оборудования.

Keywords: trans fatty acids, cottonseed oil, FTIR spectroscopy, chemometrics, thermal degradation.

Ключевые слова: трансжирные кислоты, хлопковое масло, ИК-Фурье-спектроскопия, хемометрия, термическая деградация.

Introduction

Trans fatty acids (TFAs) are a type of unsaturated fatty acids characterized by at least one double bond in the trans configuration. They are primarily found in two forms: naturally occurring TFAs in meat and dairy products from ruminant animals, and industrially produced TFAs formed during partial hydrogenation of vegetable oils. [1] Trans fatty acids (TFAs) are present in various foods, with their concentrations significantly influenced by their sources and processing methods. Industrially produced TFAs, such as those found in partially hydrogenated oils, typically range from 25-45% by weight, while processed and fried foods can contain between 0.1-10 grams of TFAs per 100 grams. Naturally occurring TFAs are present in smaller amounts in ruminant products; butter contains approximately 2-7 grams per 100 grams, and whole milk has about 0.07-0.1 grams per 100 grams. Globally, TFA intake ranges widely, averaging from 0.3% to 4.2% of total energy intake across different countries, highlighting the importance of monitoring dietary sources for effective public health interventions. [2,3]

The consumption of TFAs has been linked to several adverse health effects. Notably, TFAs increase low-density lipoprotein (LDL) cholesterol levels while decreasing high-density lipoprotein (HDL) cholesterol levels, thereby elevating the risk of coronary heart disease (CHD). Additionally, high TFA intake is associated with systemic inflammation, insulin resistance, and an increased risk of type 2 diabetes. The World Health Organization (WHO) estimates that more than 278,000 deaths annually are attributable to the intake of industrially produced TFAs. [4]

TFAs originate from both natural and industrial sources. Naturally occurring TFAs are present in small amounts in meat and dairy products from ruminant animals. In contrast, industrial TFAs are produced during the partial hydrogenation of vegetable oils, a process that solidifies oils to enhance shelf life and stability. These industrial TFAs are commonly found in processed foods such as baked goods, fried foods, and margarine. [1,5,6,7]

In response to the health risks associated with TFAs, many countries have implemented regulations to limit their presence in food products. For instance, the European Union has set a maximum limit of 2 grams of industrial TFAs per 100 grams of fat in foods intended for the final consumer. Similarly, the United States Food and Drug Administration (FDA) has declared that partially hydrogenated oils are no longer ‘generally recognized as safe,’ leading to their removal from the list of approved food additives. The WHO has also launched the REPLACE initiative, aiming to eliminate industrially produced TFAs from the global food supply.

Accurate detection and quantification of TFAs in food products are crucial for regulatory compliance and public health. Several analytical methods have been developed for this purpose.  Fourier-transform infrared spectroscopy (FTIR) is a rapid, non-destructive analytical technique widely used for screening trans fatty acids in edible fats and oils. FTIR operates by measuring the infrared absorbance of characteristic functional groups in fatty acids. The presence of trans double bonds generates unique absorption bands near 966 cm⁻¹, enabling quantification of TFA content. FTIR is advantageous due to its speed, minimal sample preparation, and suitability for high-throughput screening, though it is often complemented by chromatographic methods to enhance specificity and accuracy. [8,9] Gas chromatography-mass spectrometry (GC-MS) is considered the gold standard for the precise identification and quantification of individual trans fatty acid isomers. This method involves converting fatty acids into volatile fatty acid methyl esters (FAMEs) through derivatization, typically by methylation, followed by separation on a capillary gas chromatography column. The separated FAMEs are subsequently identified and quantified by mass spectrometry, offering high sensitivity, specificity, and detailed structural information about the fatty acids. GC-MS provides reliable results for regulatory compliance, nutritional labeling, and research on dietary fats. [10,11]

Both FTIR and GC-MS are valuable analytical tools, with FTIR commonly employed for routine screening and rapid assessments, while GC-MS is utilized for detailed, definitive characterization of trans fatty acids in various food matrices.

In this study, we developed a workflow integrating FTIR spectroscopy with chemometrics to quantify major fatty acids and potential additives in cottonseed oil samples. The method was applied to three commonly used samples collected in 2022 from Uzbekistan. In FTIR spectra, nearly 20 spectral bands corresponding to functional groups of oil components were annotated, including a characteristic peak at 966 cm⁻¹, specific to trans double bonds, which was used to construct a calibration model for TFA quantification.

Materials and Methods

Three brands of commonly used cottonseed oils in Uzbekistan were selected for this study. These oils were obtained from local markets and include locally produced brands of oils. The samples were stored at 22°C in sealed, dark glass containers prior to analysis to prevent pre-experimental oxidation and then used as control samples.

To assess the effect of temperature on the chemical properties of these vegetable oils, controlled heating experiments were conducted at three different temperatures: 100°C, 180°C, and 250°C. These temperatures were chosen to reflect common culinary and industrial heating conditions. Each oil sample (50 mL) was placed in a heat-resistant glass beaker and subjected to heating on a thermostatically controlled hotplate.

The oil samples were heated for 0, 10, 60, and 180 minutes at each temperature to simulate short- and long-term thermal exposure. To prevent local overheating, the samples were stirred continuously. After heating, samples were immediately cooled to room temperature in an ice bath to halt further chemical reactions.

Each oil sample was collected at the predefined time intervals and stored at -80℃ in amber glass vials to minimize light-induced oxidation. The samples were then filtered through a 0.45 µm membrane filter to remove any potential particulate matter before chemical analysis.

An aliquot of oil was transferred onto the diamond surface of a Bruker Vertex 80 spectrometer for measurement. The spectrometer was equipped with a deuterated triglycine sulfate (DTGS) detector, a potassium bromide (KBr) beam splitter, and a single-reflection Platinum ATR diamond accessory. The spectra were recorded in the range of 400-4000 cm-1 with a resolution of 4 cm-1 by co-adding 128 scans per sample. Each sample was measured in triplicate to assess relative standard deviation of measurements. Prior to each measurement, the ATR diamond surface was cleaned with ethanol and dried thoroughly with a clean wipe before recording the background interferogram.

Principal component analysis (PCA) and linear regression analysis were used to evaluate the results. [12,13] FTIR spectra were preprocessed using a Savitzky-Golay filter, followed by mean centering prior to performing PCA. [14] For linear regression to quantify TFA%, the FTIR spectra were preprocessed with a Savitzky-Golay filter, followed by calculating the negative second-order derivative. All computations were performed in Python (v3.12.4).

Results and Discussions

The determination of trans fatty acid (TFA) content in oil samples was performed using the FTIR spectral region centered at 966 cm⁻¹, which is a well-established marker for trans double bonds (C=C stretching) in unsaturated fatty acids. [15] To enhance spectral clarity and eliminate baseline shifts, a negative second-order derivative preprocessing technique was applied. [16] The resulting derivative spectra (Figure 1) displayed baseline-corrected profiles, with the signal intensity at 966 cm⁻¹ showing a strong correlation with TFA levels in the calibration standards.

Figure 1. Raw and baseline corrected FTIR spectra of oil samples in the trans double bond region (900-990 cm⁻¹). The left panel shows the unprocessed absorbance spectra, while the right panel presents the spectra after baseline correction, highlighting the diagnostic trans-band near 966 cm⁻¹

A calibration curve (Figure 2) was generated based on the derivative signal at 966 cm⁻¹, using a series of five spiked standards prepared by adding known amounts of elaidic acid to pure triolein, with final concentrations ranging from 0.2% to 10% (w/w). The linear regression analysis revealed an excellent correlation (R² = 0.99) between the absorbance at 966 cm⁻¹ and the known TFA concentrations. This model was then used to quantify TFA content in the analyzed oil samples. The root mean square error (RMSE) for TFA quantification was as low as 0.9%, indicating high precision and reliability of the method.

Figure 2. Calibration curve for the quantification of trans fatty acids using FTIR spectroscopy. The measured signal at the trans-specific absorption band (~966 cm⁻¹) is plotted against known concentrations of trans fatty acids, showing excellent linearity (R² = 0.99)

Figure 3. Variation of trans fatty acid (TFA) content (%) in three different cottonseed oil samples across increasing temperature and time. The table below the figure shows the average values of sample replicates. Point 1 - 22℃, 0 minutes; point 2 - 100℃, 0 minutes; point 3 - 180℃, 10 minutes; point 4 - 180℃, 60 minutes; point 5 - 180℃, 180 minutes

Figure 4. Variation of trans fatty acid (TFA) content (%) in three different cottonseed oil samples across increasing temperature and time. The table below the figure shows the average values of sample replicates. Point 1 - 22℃, 0 minutes; point 2 - 100℃, 0 minutes; point 3 - 250℃, 10 minutes; point 4 - 250℃, 60 minutes; point 5 - 250℃, 180 minutes

At room temperature (22 °C, 0 minutes), the initial TFA content in all samples ranged from 0.13% to 0.21%, reflecting naturally occurring TFAs or trace amounts formed during prior processing. Heating at 100 °C (0 minutes) did not substantially alter TFA levels, suggesting minimal isomerization under moderate thermal stress. However, once the temperature increased to 180 °C and above, a time-dependent increase in TFA formation became evident.

In the experiment shown in Figure 3, heating oils at 180 °C caused a gradual rise in TFA content from 0.26%-0.31% (10 minutes) to 0.85%-0.94% (180 minutes). In contrast, Figure 4 demonstrates that treatment at 250 °C resulted in a much sharper TFA increase, with final values reaching 3.19%-3.27% after 180 minutes. These findings indicate a nonlinear relationship between TFA formation and thermal input, with significantly enhanced isomerization kinetics occurring above 200 °C.

The observed patterns align with prior studies reporting that trans isomerization of unsaturated fatty acids, particularly linoleic and linolenic acids, is induced by high thermal energy and prolonged exposure time. Under such conditions, cis-configured double bonds in polyunsaturated fatty acids (PUFAs) are converted into thermodynamically more stable trans-isomers via radical-mediated and pericyclic mechanisms, especially in the absence of sufficient antioxidants. [17,18]

Moreover, the consistent behavior of all three cottonseed oil samples across both experiments suggests that the fatty acid matrix of refined cottonseed oils responds uniformly to thermal stress, with minor deviations likely due to natural variation in antioxidant content or residual minor compounds. This reinforces the conclusion that temperature and heating duration are the dominant factors influencing TFA generation, while initial oil composition plays a secondary role.

The health implications of these results are significant. Trans fatty acids are well-documented contributors to cardiovascular disease, systemic inflammation, and insulin resistance. [19,20,21] The fact that TFA content in oils can rise from <0.3% to over 3% under common frying conditions underscores the potential dietary hazard associated with repeated high-temperature oil use, particularly in household and food industry settings.

These results support global recommendations to limit TFA intake and emphasize the importance of regulating oil heating practices, promoting the use of thermally stable oils, and monitoring processed food products for TFA content.

Conclusion

This study demonstrates the efficacy of FTIR spectroscopy coupled with chemometric modeling for the quantification of trans fatty acids in cottonseed oils subjected to thermal treatment. The method is rapid, reproducible, and well-suited for routine monitoring of oil quality. Results clearly show that prolonged heating—especially at temperatures above 200 °C—leads to a substantial increase in TFA content, with levels reaching over 3% at 250 °C after 180 minutes. This trend is consistent across all tested oil samples, suggesting that refined cottonseed oil undergoes similar isomerization pathways regardless of brand-specific differences. Given the established links between dietary TFAs and cardiovascular risk, the findings reinforce the necessity of regulating heating practices and minimizing repeated oil usage in both domestic and industrial settings. The proposed FTIR-based strategy offers a viable, cost-effective alternative to GC-MS for TFA surveillance in edible oils and may be extended to broader food safety applications.

References:

1. U. Mavlanov et al., ‘The effects of industrial processing and home cooking practices on trans-fatty acid profiles of vegetable oils’, Food Chem., Vol. 469. - PP. 142571, Mar. 2025.  [Electronic resource] URL: doi: 10.1016/j.foodchem.2024.142571.
2. V. Dhaka, N. Gulia, K. S. Ahlawat, and B. S. Khatkar, ‘Trans fats—sources, health risks and alternative approach - A review’, J. Food Sci. Technol., Vol. 48, No. 5. – PP. 534-541, Oct. 2011. [Electronic resource] URL: doi: 10.1007/s13197-010-0225-8.
3. J. Wanders, P. L. Zock, and I. A. Brouwer, ‘Trans Fat Intake and Its Dietary Sources in General Populations Worldwide: A Systematic Review’, Nutrients, Vol. 9, No. 8, Art. No. 8, Aug. 2017. [Electronic resource] URL: doi: 10.3390/nu9080840.
4. D. Pipoyan et al., ‘The Effect of Trans Fatty Acids on Human Health: Regulation and Consumption Patterns’, Foods, Vol. 10, No. 10. – PP. 24-52, Oct. 2021. [Electronic resource] URL:  doi: 10.3390/foods10102452.
5. A.-B. Oteng and S. Kersten, ‘Mechanisms of Action of trans Fatty Acids’, Adv. Nutr., Vol. 11, No. 3. – PP. 697-708, May 2020. [Electronic resource] URL: doi: 10.1093/advances/nmz125.
6. S. Bhardwaj, S. J. Passi, and A. Misra, ‘Overview of trans fatty acids: Biochemistry and health effects’, Diabetes Metab. Syndr. Clin. Res. Rev., Vol. 5, No. 3. PP. 161-164, Jul. 2011. [Electronic resource] URL: doi: 10.1016/j.dsx.2012.03.002.
7. S. T. Alexandra Krettek and B. Göran Bondjers (Nordic School of Public Health through MILIEU Ltd., ‘Trans Fatty Acids and Health: A Review of Health Hazards and Existing Legislation’, Nov. 2008, Accessed: Apr. 16, 2025. [Online]. [Electronic resource] URL: https://policycommons.net/artifacts/1341417/trans-fatty-acids-and-health/1953384/
8. S. T. H. Sherazi, A. Kandhro, S. A. Mahesar, M. I. Bhanger, M. Y. Talpur, and S. Arain, ‘Application of transmission FT-IR spectroscopy for the trans-fat determination in the industrially processed edible oils’, Food Chem., Vol. 114, No. 1. – PP. 323-327, May 2009. [Electronic resource] URL: doi: 10.1016/j.foodchem.2008.09.058.
9. M. M. Mossoba, C. T. Srigley, and J. K. G. Kramer, ‘Portable FTIR analyzers for the rapid determination of total trans-fat’, Lipid Technol., Vol. 27, No. 1. – PP. 11-13, Jan. 2015. [Electronic resource] URL: doi: 10.1002/lite.201400074.
10. P. Delmonte and J. I. Rader, ‘Evaluation of gas chromatographic methods for the determination of trans fat’, Anal. Bioanal. Chem., Vol. 389, No. 1. – PP. 77-85, Aug. 2007. [Electronic resource] URL: doi: 10.1007/s00216-007-1392-y.
11. P.-A. Golay, F. Giuffrida, F. Dionisi, and F. Destaillats, ‘Streamlined Methods for the Resolution and Quantification of Fatty Acids Including Trans Fatty Acid Isomers in Food Products by Gas Chromatography’, J. AOAC Int., Vol. 92, No. 5. – PP. 1301-1309, Sep. 2009. [Electronic resource] URL: doi: 10.1093/jaoac/92.5.1301.
12. H. Hotelling, ‘Analysis of a complex of statistical variables into principal components.’, J. Educ. Psychol., Vol. 24, No. 6. – PP. 417-441, Sep. 1933. [Electronic resource] URL: doi: 10.1037/h0071325.
13. F. Galton, ‘Regression Towards Mediocrity in Hereditary Stature.’, J. Anthropol. Inst. G. B. Irel., Vol. 15. – PP. 246, 1886. [Electronic resource] URL: doi: 10.2307/2841583.
14. Abraham. Savitzky and M. J. E. Golay, ‘Smoothing and Differentiation of Data by Simplified Least Squares Procedures.’, Anal. Chem., Vol. 36, No. 8. – PP. 1627-1639, Jul. 1964. [Electronic resource] URL: doi: 10.1021/ac60214a047.
15. M. M. Mossoba, M. P. Yurawecz, and R. E. McDonald, ‘Rapid determination of the total trans content of neat, hydrogenated oils by attenuated total reflection spectroscopy’, J. Am. Oil Chem. Soc., Vol. 73, No. 8. – PP. 1003-1009, Aug. 1996. [Electronic resource] URL: doi: 10.1007/BF02523408.
16. M. M. Mossoba et al., ‘Regulatory Infrared Spectroscopic Method for the Rapid Determination of Total Isolated Trans Fat: A Collaborative Study’, J. Am. Oil Chem. Soc., Vol. 88, No. 1. – PP. 39-46, 2011. [Electronic resource] URL: doi: 10.1007/s11746-010-1648-0.
17. F. Destaillats, J. P. Trottier, J. M. G. Galvez, and P. Angers, ‘Analysis of α-Linolenic Acid Biohydrogenation Intermediates in Milk Fat with Emphasis on Conjugated Linolenic Acids’, J. Dairy Sci., Vol. 88, No. 9. – PP. 3231-3239, Sep. 2005. [Electronic resource] URL: doi: 10.3168/jds. S0022-0302(05)73006-X.
18. H. t. Osborn and C. c. Akoh, ‘Structured Lipids-Novel Fats with Medical, Nutraceutical, and Food Applications’, Compr. Rev. Food Sci. Food Saf., Vol. 1, No. 3. – PP. 110-120, 2002. [Electronic resource] URL: doi: 10.1111/j.1541-4337. 2002.tb00010. x.
19. D. Mozaffarian, M. B. Katan, A. Ascherio, M. J. Stampfer, and W. C. Willett, ‘Trans Fatty Acids and Cardiovascular Disease’, N. Engl. J. Med., Vol. 354, No. 15. – PP. 1601-1613, Apr. 2006. [Electronic resource] URL: doi: 10.1056/NEJMra054035.
20. W. H. Organization and I. A. Brouwer, Effect of trans-fatty acid intake on blood lipids and lipoproteins: a systematic review and meta-regression analysis. World Health Organization, 2016. Accessed: Apr. 17, 2025. [Online]. [Electronic resource] URL: https://iris.who.int/handle/10665/246109
21. Sharofiddin N. et al. Chemical transformations of vegetable oils during cooking: implications for cardiovascular diseases and type 2 diabetes //Universum: chemistry and biology. – 2025. – Vol. 2. – №. 2 (128). – PP. 19-25.